Scholarlink

This study aims to develop predictive models for key performance indicators of SRDI (Specialized, Refined, Differentiated, and Innovative) enterprises, namely profitability, R&D capabilities, and governance levels, to better understand and support their high-quality development. Utilizing a dataset of 83 A-share listed companies recognized as national-level SRDI “Little Giant” enterprises, we compare traditional time series models (Mean and ARIMA) with advanced deep learning models (LSTM, GRU, and a hybrid LSTM-GRU model). Our results demonstrate that the LSTM-GRU hybrid model significantly outperforms other models. Furthermore, we applied this model to predict key indicators for SRDI enterprises and provided management recommendations based on the predictions.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A New LSTM-GRU Hybrid Model for Forecasting the Growth of SRDI Enterprises

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Lewei Hu

摘要

This study aims to develop predictive models for key performance indicators of SRDI (Specialized, Refined, Differentiated, and Innovative) enterprises, namely profitability, R&D capabilities, and governance levels, to better understand and support their high-quality development. Utilizing a dataset of 83 A-share listed companies recognized as national-level SRDI “Little Giant” enterprises, we compare traditional time series models (Mean and ARIMA) with advanced deep learning models (LSTM, GRU, and a hybrid LSTM-GRU model). Our results demonstrate that the LSTM-GRU hybrid model significantly outperforms other models. Furthermore, we applied this model to predict key indicators for SRDI enterprises and provided management recommendations based on the predictions.